Voluntary error reporting systems are an important part of safety improvement programs, but difficulty in analyzing error reports has limited their utility. This study described the development of a machine learning algorithm to analyze free-text data in incident reports. The algorithm proved to be accurate in classifying events when compared to manual review.

Using ventilator-associated adverse events as a test case, this study illustrates how the public–private partnership for the promotion of patient safety concept can augment understanding of patient safety issues. The group successfully developed a common taxonomy for evaluating adverse events from three different reporting systems.

Patient safety reporting systems are commonplace in most organizations as a tool to identify, track, and potentially prevent adverse events despite their known limitations. Heparin is a high-risk medication that frequently generates incident reports, and significant efforts have been established to ensure its safe use. This study reviewed reported heparin errors from three large patient safety reporting systems—MEDMARX, the Pennsylvania Patient Safety Authority, and the University Health Consortium (an alliance of academic medical centers)—to capture events from more than 1000 organizations. Of the 300,000 medication events reported, approximately 4% involved heparin products, with the administration phase being the most frequently cited. As this was the first study to combine three large sources of reporting data for a single process, the authors point out the consistent patterns detected, suggesting diminishing returns from aggregating reports around common events.

Patients are increasingly being asked to assume a role in ensuring their own safety. Both AHRQ's "20 Tips to Help Prevent Medical Errors" and the Joint Commission's "Speak Up" program to ensure surgical safety recommend that patients engage in specific safety practices, such as maintaining a list of their medications and asking health care workers if they have washed their hands. This survey assessed the willingness of patients to carry out these practices. Patients were much less likely to engage in behaviors that required them to challenge providers (such as checking for handwashing compliance) than less confrontational practices (such as following up on test results). The study's findings are similar to a prior AHRQ-funded study of patients recently discharged from the hospital.

This article reviews core principles of patient safety that health care leadership should know in order to prevent errors and foster safety.

This article describes the safety efforts of the Pennsylvania Patient Safety Authority, which includes their incident reporting system (PA-PSRS) and publication of reporting trends in Patient Safety Advisory.